Over the past two decades, the molecular genetic analyses of human cancers have led to unprecedented insights into these diseases. This is especially apparent in human colorectal cancers where a series of somatic mutations that drive the process has been delineated, the genetic bases of several inherited predispositions to cancer have been defined, and detailed expression analyses have identified a myriad of expression changes associated with neoplastic growth. The goals of this project are to apply this new knowledge to the improved management of human colorectal cancer. The aims of this project can be divided into efforts designed to use somatic mutations (Aim 1), germline mutations (Aim 2), and gene expression analyses (Aim 3) to improve early detection, prevention and management of human colorectal tumors. The objectives of Aim 1 are to use somatic mutations to detect early colorectal tumors when they are still curable and to categorize cancers according to their genetic signatures. The objective of Aim 2 is to improve our ability to identify individuals with inherited predispositions to colon cancer. The objective of Aim 3 is to use gene expression analysis to identify tumor markers that may be useful for early diagnosis. This project will thereby combine efforts originally initiated in current SPORE projects 0012 and 0014. Though the aims may seem diverse, their inclusion in a single project benefits from the sharing of key personnel and patient resources. Likewise, the aims may seem ambitious, but each is supported by substantial preliminary data.